The Argument for School Closures This Fall Simply Does Not Add Up

We have undergone the worst pandemic since the 1918 Spanish Flu. With it has come the worst economic downturn since the Great Depression. The United States has also experienced the worst social unrest since 1968. While we adapt to a "new norm," there has been considerable debate on the policy responses to COVID-19. In an unprecedented move, we severely limited movement by closing down large swathes of the economy with lockdowns. I have made it no secret that I have thought that the lockdowns were a terrible idea (see here, here, and here). There have been other non-pharmaceutical interventions [NPI] outside of lockdowns that have been used to flatten the curve. Aside from social distancing, face masks are an example of such an NPI. I have made an argument for a temporary, limited face mask mandate (see here and here). 

With the new academic year quickly approaching, another COVID-related NPI has gained attention: school closures. This past spring, there were a number of schools that closed down in attempts to limit the spread of COVID-19. Other schools allowed for children to play freely with one another, whereas other school remained open with considerable social distancing provisions. Much like with the lockdowns, the school closures have become one large social experiment. This brings us to the main question for this fall: Should schools be closed for the upcoming academic year? If not, which precautionary measures make most sense to implement while keeping schools open this year? 

Assessing Benefits of School Closures: Are schools hotspots for COVID-19?

The main purpose of keeping schools closed is to stop the spread of COVID-19. There is some intuition that children would be vectors of COVID-19. They are less likely to keep social distancing, basic levels of hygiene, or wearing their face masks all the time. Plus, children are in contact with parents, classmates, parents, and other school staff. The amount of people combined with apparent lack of adherence to NPI protocols would make children likely culprits of spreading COVID-19. Let's ask some key questions.

What is the COVID-19 fatality rate for children? 

"Think of the children." This sort of political rhetoric has been used to advance multiple policies, ranging from subsidizing school lunches and banning video games to the even-stupider idea of banning adoptions from same-sex parents. So let's focus on the children for a moment. Available public health data show that children account for 22 percent of the population, but two percent of the COVID deaths in the United States. Looking at international data, the countries of Spain, China, and Italy (collected by Oxford University) show that the case fatality rate for those under 20 is near-zero. 

Are children super-transmitters of COVID-19? 

The publication Pediatrics concluded that there have not been that many child-to-adult transmissions of COVID-19 (Lee and Raszka, 2020). Other studies show that children are one-third to one-half less likely to transmit COVID-19 (Bi et al., 2020; Li et al., 2020; Mizumoto et al., 2020; Zhang et al., 2020).

How will this affect teachers and other school staff?

At the same time, schools do not solely consist of children. There are teachers and other school staff that work in schools, many of whom whose age alone puts them in a high-risk category. While there is not a lot of data on teachers and school staff, there is one French study with 46 teachers showing there has not been a single case of transmission (Fontanet et al., 2020). Also, the average age of teachers in the U.S. is 42.4 years (National Center of Education Statistics), 81 percent of whom are under 55 years old. We should take precautions for teachers (especially older and/or more immunocompromised teachers), but teachers are, on average, below the age of being considered high-risk for COVID-19. 

Have school re-openings resulted in spikes of COVID-19 cases?

One good way to ask about the overall risk is to see if school openings or re-openings meant major spikes in COVID-19. Israeli schools experienced a second round of school closures. There was an outbreak at a school in New Zealand. At the same time, such outbreaks are exceptions, not the norm. Preliminary data from 22 nations in the European Union indicate that much of the European Union did not experience a spike in COVID-19 cases. Let's take a look at some of the EU countries. After opening up their schools, the Dutch government found that there were no children infected and very few school employees. Folkhälsomyndigheten, which is the Swedish Health Ministry, compared its school policy with that of Finland. Sweden kept its schools open, whereas Finland closed them. The Minister concluded there was no discernible difference in terms of COVID transmission (Carlson, 2020). Although Denmark gave Sweden some grief, Denmark ended up reopening their schools. And guess what? It didn't worsen the COVID outbreak in Denmark (Reuters). Ditto with Ireland (Heavey et al., 2020), Austria, Finland, and Taiwan (Esposito and Principi, 2020), as well as certain reported regions in France (Danis et al., 2020) and Australia. 

Will keeping schools open result in an increase of COVID-related deaths?

There is a bit of conflicting information on the answer to that question. A systematic review in The Lancet estimated that school closures would lower deaths by 2 to 4 percent, which is still something, but less effective compared to other NPIs (Viner et al., 2020). There are some studies that suggest the opposite. A study from The Lancet states that "school closures could lead to a greater number of deaths than they prevent" (Bayham and Fenichel, 2020). Another study from Health Affairs made a similar suggestion that not only are school closures ineffective, but they could cause greater death (Courtemanche et al., 2020). 

Costs of School Closures

Given what we have seen in schools thus far (see previous sub-section), I am inclined to think that it will not have a significant effect on the overall COVID death count. Even Kevin Drum at Mother Jones, a Left-of-center magazine, said that the current evidence on school closures have "a) little effect, and b) are probably nowhere near worth the tremendous impact they have on both parents and kids." Speaking of which, what is the impact that school closures have? 

• Remote learning means lower student achievement. Measuring RIT scores, the Brookings Institution found a loss in learning projection (Soland et al., 2020). It is estimated that the school closures last spring translated  into only having 70 percent of the learning gains one would normally have. It is even worse for mathematics, with less than 50 percent. Another study goes as far as suggesting that the school closure in spring results in losing at least nine months of educational growth if the school closures continue (Christakis, 2020).
• Economic loss and future earnings. The Brookings Institution has conducted some work on the matter. Their preliminary estimate puts the loss of earnings of $1,337 per year per student (Psacharopoulos et al., 2020). An older study from the Brookings Institution estimated that a 12-week school closure would mean a loss of over $120 billion lost in GDP [in 2008 dollars] (Lempel et al., 2009). 
• The achievement gap and its effects on race and income. Survey data from Pew Research found that children in low-income households were less likely to be able to complete homework at home. According to a survey from The Education Trust, 76% of African-American parents and 82% of Hispanic parents are concerned they do not have the adequate resources to keep their children on track. This has direct impact on being able to achieve in school in the long-run. Consulting firm McKinsey released a report on the achievement gap in the COVID-19 pandemic (Dorn et al., 2020). In addition to the $110 billion in annual earnings lost that McKinsey estimates, there is also a greater achievement gap as a result. While white students are to expect a 1.6 percent reduction in future earnings, black students and Hispanic students are expected to have a 3.3 percent and 3.0 percent reduction, respectively. 
• GDP and costs of parent absenteeism. Children staying home means that parents are more likely to use time off to take care of their children. NYU epidemiology professor Jeffrey Epstein calculated that a month-long nationwide closure would mean $50 billion in lost productivity due to absenteeism.  
• Mental health concerns. Students in China were found to have increased anxiety and depression (Xi et al., 2020). A study from The Lancet concluded that students with preexisting mental health conditions had those conditions exacerbated (Lee, 2020). Survey data from Gallup finds three in ten parents saying that their child's mental health is suffering. This says nothing of what the parents are trying to juggle between work and having children at home. 
• Child abuse and neglect. As the Heritage Foundation points out, 3.5 million children came into contact with Child Protective Services in 2018. This is alarming considering that parents are the perpetrators in 92 percent of child maltreatment cases, according to an Administration of Children and Families study. The Brookings Institution details in its report how COVID-induced school closures contribute to the high likelihood of there being a considerable increase in child abuse and neglect. 

Postscript

We do not have the luxury of living in a world without risks. Whatever we choose entails at least some risk, which is why we need to weigh the potential benefits of school closures against their costs. However, when we do so using the available evidence, the choice becomes clearer and clearer. A chief scientist at the WHO, Soumya Swaminathin, said in May that children are "less capable" of spreading the virus and are at "very low risk" of the illness. The accumulating data are showing that children are not a main driver of COVID transmission (Ludvigsson, 2020). The Royal College of Pediatrics and Child Health made an emphatic statement by saying how this risk is unprecedented and how it could scar the life chances of a generation of young people. As a study from The Lancet concluded, the evidence base for school closures is weak, whereas the costs are high (Viner et al., 2020). To quote the American Academy of Pediatrics, "all policy considerations for the coming year should start with a goal of having students physically present in school." While particularly hard-hit districts will have to reassess in light of exigent circumstances, the default should be to open schools for the upcoming academic year.

The American Academy of Pediatrics provides detailed guidelines on how to do so, but hand-washing, screening, physical distancing, environmental cleaning, proper ventilation, cancelling pep rallies and other large gatherings, staggered scheduling, and offering the possibility of high-risk employees either remote teaching or the semester off would go a long way in minimizing COVID-19 spread because truth be told, we cannot afford another semester without in-person learning. 

9-13-2020 Addendum: In case you needed more costs to the school closures. One paper from the OECD found that the school closures will reduce a student's lifetime income by 3 percent (Hanushek and Woessmann, 2020). Another paper from the National Bureau of Economic Research shows how school closures stunt skills attainment (Fuchs-Schündeln et al., 2020).